Since the wide-spread success of the polymerase chain reaction (PCR) technology, microbiology laboratories are waiting for the application of molecular biology to routine microbiology. This has been held back by an inherent and fundamental problem of molecular biology. Because of its precision, you need to know which tools (probes) to choose. The prerequisite is, that a request has to be specified with respect to organisms to be detected. In clinical samples, however, you do not know which of the over 2000 clinically relevant pathogens is the causative agent of an infection. A rational approach solves the problem                Focus must be made on 95% of problem causing organisms        If it is known where the sample was taken, and clinical data is present, the number of organisms can be reduced to between 2 and 16.        The number of organisms to cover the 95-percentile in most clinical samples is in the order of 100        
This rationale makes it economically feasible to run a DNA-probe based assay on a routine basis.
Grouping of micro-organisms and the very rapid testing for presence/absence of specific or a range of micro-organisms is also of relevance in other fields of microbiological testing: Blood banks, Pharmaceutical industry, Cosmetic industry and the Food industry. Frequently the same organisms are of relevance throughout the disciplines and reaction conditions therefor need to be standardised for all probes.
The detection of ribosomal RNA via Fluorescent in-situ Hybridisation (FISH) or utilising chip technology represents an efficient way of utilising sensitivity and specificity of DNA-probes without having to use an enzymatic amplification step. FISH relies on two approaches to the in-situ detection of targets generating a signal strong enough to be detected with standard measuring devices such as an epifluorescence microscope:                1. Identical molecules are present within a cell in sufficient numbers to bind one specific oligonucletide or nucleotide analogue probe with one fluorophor each.        2. Large probes carrying a plurality of fluorophores e.g. labelled cosmids.        
FISH technology for the identification of micro-organisms in their respective environments is well known in the art. Application of FISH for the detection of pathogens is of especial interest to the clinical microbiology and infectiology, where FISH excels in speed and cost efficiency.
Detecting rRNA with chip technology also relieves from the necessity to amplify the target. Total rRNA is extracted from a sample and placed on a chip. Specific probes are concentrated on a small surface area and attract respective rRNA-molecules to give specific presence/absence signals. In order to make such chips economically viable they need to be used repeatedly with as little manipulations as possible. Furthermore the standardisation of probe characteristics is paramount for the generation of reproducible results.
In order to gain acceptance in a routine environment probes must be designed in such a way that all probes for one disease state can be run simultaneously under identical conditions in or on one vessel (chips, micro-fluidic devices or micro titre plates). In the design of the probes and to make the probes economically viable, it must be taken into account that one probe may be of relevance to different disease states. Therefor, not only one set of probes but all probes must work under identical hybridisation conditions. Sequence and length of the working probes must be tested accordingly.
The selection and definition of a working probe cannot be performed by simple sequence comparison and determination of a theoretical Tm-value. Depending on the algorithm applied a wide set of values are obtained giving no guidance to the choice of probe sequence suitable for standardised hybridisation conditions.
The choice of algorithm and factors influencing the quality of a probe is discussed widely in the art (1-7). Further guidance may be sought comparing sequences and actual position in the three dimensional structure of the ribosome. In an attempt to rationalise the design of probes Behrens et al (8) investigated the correlation between hybridisation sites and actual accessibility with the help of the 3 Å three dimensional model of the ribosome. Their findings demonstrated that the SDS used in in-situ procedures has a predominant denaturing effect, not captured by algorithms predicting secondary structures.
A further problem in both FISH and chip technology is that the procedure calls for a stringent wash step to remove unbound probes, requiring additional handling steps, reagents and time. The success of a hybridisation may depend largely on the skill and precision applied to the washing step. However, routine applications call for minimal steps and hands-on time, most importantly they must be independent from individual skills.
One solution to the reduction of steps would be the application of fluorescence resonance energy transfer (“FRET”) in an oligo-nucleotide or nucleotide analogue hairpin formation (molecular beacon). Several approaches to the development of beacons are known in the art and generalised descriptions to their construction are freely available (13). Beacons are widely used in real time PCR, where they anneal in solution to an increasing number of templates generated by amplifying enzymes (15). Only few attempts have been made to generate beacons for the detection/identification of bacteria on membranes (14). One successful beacon was constructed to detect E. coli in whole cells with a peptide nucleic acid (PNA) probe. The corresponding DNA-probe failed to give adequate performance (16). The production of further PNA-beacons is limited due to the poor solubility of PNA based oligonucleotides as laid down in design recommendations (17).
Patent CA 2176266/EP 0745690 gives guidance to the construction of universal stems for real time PCR (9). Surprisingly, these recommendations do not render working beacons when combined with probes designed to identify micro-organisms in-situ. Real time PCR is performed in solution while both ISH (in-situ hybridisation) and chips require fixed targets. Their thermodynamic details were not compatible with in-situ hybridisation and FRET requirements. Thus, universally working stems could not be predicted for applications with fixed targets. It was therefore necessary to empirically search for specific beacons fitting individual oligo-nucleotide or nucleotide analogues in order to accomplish a plurality of beacons working under identical ISH specifications.
In the selection of ISH-beacons care has to be taken that the stem does not hinder the delicate balance of hybridising towards RNA entwined in large protein/RNA complexes such as ribosomes. The accessibility of binding sites is widely discussed in the art and is summarised in (1).
Further limitations in the design of a beacon probe are given by the size of pores generated in the cell wall during the ISH procedure. Adding the same stem to different probes results in distinctly individual beacons. A plurality of probes already form hairpin loops and the addition of a stem does not result in a “beacon” formation. In addition, simply adding bases to form complementary pairs may increase the Tm to such an extent that the hairpin is thermodynamically preferred rather than the hybrid formation. Special stems have to be devised that pull the sequence into beacon formation while maintaining the Tm at or below that of the hybrid. The teachings with respect to the design of beacons (13) show that the increase of the stem length by one base pair increases the Tm by 5° C. and that the Tm of the stem should be 10° C. higher than the Tm of the hybridising sequence.
FISH with single microorganisms, such as bacteria, based upon specific rRNA sequences, may be difficult due to sterical hindrance of the rRNA in the ribosome. In other words, a beacon forming a hairpin may poorly anneal to the embedded rRNA target sequence.
It is therefore the subject of the present invention to provide molecular beacons which overcome the above described disadvantages at least partially. The solution provided in the present invention and preferred embodiments thereof are described in the claims.
Subject of the present invention is a nucleic acid capable of forming a hybrid with a target nucleic acid sequence and capable of forming a stem-loop structure if no hybrid is formed with the target sequence, said nucleic acid comprising    (a) a nucleic acid portion comprising            (a1) a sequence complementary to the target nucleic acid sequence,        (a2) a pair of two complementary sequences capable of forming a stem,            (b) an effector and an inhibitor, wherein the inhibitor inhibits the effector when the nucleic acid forms a stem-loop structure, and wherein the effector is active when the nucleic acid is not forming a stem-loop structure.
The nucleic acid of the present invention capable of forming a hybrid with a target nucleic acid sequence and capable of forming a stem-loop structure if no hybrid is formed with the target sequence is also referred herein as “beacon”, “molecular beacon”, “hairpin”, or “hairpin loop”, wherein the “open” form (no stem is formed) as well as the “closed” form (the beacon forms a stem) is included. The open form includes a beacon not forming a hybrid with a target sequence and a beacon forming a hybrid with the target sequence.
In particular, the two complementary sequences (a2) are flanking the sequence (a1), i.e. the first sequence (a2) is attached at the 3′ end of the sequence (a1) and the second sequence (a2) is attached at the 5′ end of the sequence (a1).
The hybrid of the sequence (a1)) with the target sequence is also referred herein as “hybrid with the cognate sequence” or as “cognate hybrid”.
In the present invention, the effector may be attached at one of the two complementary sequences capable of forming a stem, whereas the inhibitor may be attached at the other of the two complementary sequences, so that the inhibitor essentially inhibits the effector activity when a stem is formed, and that the effector is active when the hairpin is open. Preferably, the effector is attached at the 5′ end or the 3′ end of the beacon, respectively, or at a position which is 1, 2, 3, 4, or 5 nucleotides distant to the 5′ end or the 3′ end, respectively. The inhibitor is preferably attached at the other end not covered by the effector, i.e. at the 3′ end or the 5′ end, respectively, or at a position which is 1, 2, 3, 4, or 5 nucleotides distant to the 3′ end or the 5′ end, respectively.
The design of the hairpin loops disclosed herein therefore differs fundamentally from beacons well known in the art.
Hybridisation of the beacon of the present invention with target sequence may take place under conditions where the loop is open. A beacon which is not forming a stem when hybridizing is capable of annealing to a target rRNA sequence, for instance, and can therefor achieve successful hybridisation.
This goal is for instance achieved by a Tm of the beacon (i.e. the Tm of the stem) which is essentially equal to or lower than the Tm of the cognate hybrid (i.e. the hybrid of the beacon with the target sequence). Thus, hybridisation with the target sequence takes place when the stem is open, for instance if hybridisation takes place under essentially Mg2+ free conditions.
“Essentially equal Tm” of the cognate hybrid and the stem of the beacon refers to melting temperatures differing in less than 5° C., preferably less than 3° C., more preferably less than 2° C., more preferably less than 1° C., more preferably less than 0.5° C., even more preferably less than 0.2° C., most preferably less than 0.1° C.
In order to achieve an inhibition of the effector by the inhibitor, both of which form part of the beacon, in those beacon molecules not hybridising with the target sequence, stem formation must be induced after the hybridisation reaction. This may for instance be achieved by a beacon having a ΔG<0, so the hairpin will form spontaneously. Further, stem formation may be introduced by washing with a Mg2+ containing buffer as described herein.
In particular, the hairpin loops are constructed in such a way that under standardised hybridisation conditions (e.g. under essentially Mg2+ free conditions) the beacon stem is open so that possible sterical limitations do not hinder the hybridisation process. For instance, sterical limitations may be present when the target sequence is a rRNA sequence. If the effector is a fluorophor, the fluorophor will not be quenched by the close proximity of ribosomal proteins.
Suitable conditions for induction of stem formation after hybridisation include an Mg2+ containing buffer, for instance containing about 1 to about 20 mM Mg2+, more particular about 5 to about 15 mM Mg2+, even more particular about 8 to about 12 mM Mg2+, most particular about 10 mM Mg2+. The buffer may have a pH>8.
Furthermore, the beacons function in their entirety and cannot be dissected into stem and loop as nearest neighbour and stacking effect have a profound influence in their thermodynamic properties. Preferred beacons of the present invention are summarised in Table 1. They clearly show that the preferred stem sequence is independent from the ΔG, Tm, GC content or length of the sequence chosen to identify a species.
In the present invention, the thermodynamic specifications for the individual construction of beacons suitable for standardised conditions are set: The Gibbs energy (ΔG) for the formation of the beacon has to be designed in such a way that                The beacon will form spontaneously (ΔG<0) in the absence of a cognate target sequence under hybridisation conditions.        The ΔG of the cognate hybrid is significantly lower (i.e. is more negative) than the ΔG of the beacon.        The respective ΔG of the beacon is lower than a mismatch or non-cognate sequence.        The Tm for the formation of the beacon has to be designed in such a way that the Tm of the beacon is lower than or essentially at the Tm of the hybrid.        
It is preferred that the ΔG of the cognate hybrid is in the range of about −17 to about −25 kcal/mol, preferably about −18 to about −24 kcal/mol, more preferably about −19 to about −23 kcal/mol, most preferably about −20 to about −22 kcal/mol under hybridisation conditions.
It is also preferred that the ΔG of the cognate hybrids under hybridisation conditions do not vary more than 5 kcal/mol, preferably no more than 3 kcal/mol, more preferably 2 kcal/mol and most preferably 1 kcal/mol.
Occasionally cognate sequences may form spontaneous hairpin loops, where one arm only needs to be supplemented to achieve the beacon formation. If the target sequence is a rRNA sequence, this, however renders the effector, e.g. the fluorophor, in very close proximity to potentially quenching proteins of the ribosome. In a preferred configuration the stem is extended. In order to conform with said thermodynamic specifications as described herein even with an extended stem a method was devised to keep both the Tm and ΔG within the specifications. According to the present invention, this can be achieved by the introduction of at least one non-matched nucleotide or nucleotide analogue. In the present invention, introduction of at least one non-matched nucleotide may be enhanced by the introduction of an additional nucleotide or nucleotide analogue, so that the two complementary sequences have a different length, and the stem becomes “bended” (see for example position 36 in SEQ ID NO:1), or/and may be achieved by a replacement of a matching nucleotide or nucleotide analogue by a non-matching nucleotide or nucleotide analogue (see for example position 5 in SEQ ID NO: 7). Thus, in the present invention, the “complementary sequences capable of forming a stem” may also include at least one non-matched nucleotide, preferably 1, 2, 3, 4 or 5 non-matched nucleotides.
As can be seen from Table 2 none of the sequences disclosed here could be devised as PNA-beacons due to the said limitations in the construction of PNA-oligonucleotides. The major limitation being in the oligonucleotide length required to have both sufficient specificity and a stem length sufficient to ensure the re-folding of the loop when not hybridised. It is therefore necessary to devise DNA-beacons that are able to hybridise with sufficient affinity and speed to enable the in-situ identification of micro-organisms.
The beacon of the present invention is not a PNA beacon. The backbone of the beacon is preferably a nucleic acid backbone. The beacon may comprise a nucleic acid analogue such as a deoxyribonucleotide analogue or a ribonucleotide analogue in the nucleic acid portion or/and in the linker if a linker is present. This analogue is preferably a nucleotide analogue modified at the sugar moiety, the base or/and the phosphate groups. The nucleotide analogue is preferably not a PNA building block.
Following the said 95-percentile in clinical samples, pathogens can be grouped into disease related groups. Probes towards these organisms must work simultaneously under the said conditions, especially if all probes are to be utilised on one chip. The chip application calls for a stringent standardisation of both the cognate and stem characteristics. If a combination of more than one probe is employed, i.e. at least two probes, all probes have to be designed to work on the same slide/chip simultaneously.
Another subject of the present invention is a combination comprising at least 2, preferably at least 10, at least 20, at least 30, at least 40, or at least 50 beacons. The combination may comprise but is not limited to all of the beacons of Table 1, preferably at the maximum 100, at the maximum 80, at the maximum 70, at the maximum 60, at the maximum 50, at the maximum 40, at the maximum 30 or at the maximum 20 beacons.
In a combination of the present invention, the beacons may have the same or different target sequences. It is preferred that the target sequences of individual beacons are different.
In a combination of beacons of the present invention, the ΔG difference of the individual beacons of the hybrid of the sequences of (a2) or/and the hybrid of the sequence of (a1) with a target sequence may be at the maximum about 4 kcal/mol, preferably at the maximum about 3 kcal/mol, more preferably at the maximum about 2 kcal/mol, and most preferably at the maximum about 1 kcal/mol with respect to the cognate sequence.
In a combination, the Tm values of individual beacons with respect to its respective cognate sequence may differ at the maximum by about 3° C., preferably at the maximum about 2° C., more preferably at the maximum about 1° C.
It is preferred that in the combination of the present invention the individual nucleic acids function uniformly. “Functioning uniformly” means that successful hybridisation can be achieved with different nucleic acids probes of the present invention under the same hybridisation conditions, for instance under standardised hybridisation conditions. In other words, uniformly functioning nucleic acids of the present invention do not require individual optimisation of the hybridisation conditions.
Depending on the disease state certain pathogens most frequently are the causative agents and can thus be compiled into diagnostic groups. Addition or omission of certain pathogens may be required depending on regional epidemiology in order to reach the 95-percentile. The preferred listing of Table 1 covers the requirements of Europe and most of North America.
Yet another aspect of the present invention is a kit or chip which may contain at least two beacons of Table 1 required to detect the listed organisms optionally together with the required hybridisation reagents, Preferably, the chip or kit contains at least 10, at least 20, at least 30, at least 40, or at least 50 beacons. The kit or chip may contain at the maximum all of the beacons of Table 1, preferably at the maximum 100, at the maximum 80, at the maximum 70, at the maximum 60, at the maximum 50, at the maximum 40, at the maximum 30 or at the maximum 20 beacons.
List of groupings and resulting kits for the detection, enumeration and identification of the listed organisms is compiled in Table 1.
The beacons can be applied to assays designed to be performed in tubes, microtitre plates, filtered microtitre wells, slides and chips. The detection can be made with fluorescence, time resolved fluorescence, with a plurality of fluorophores and utilising electrochemical enzymes.
In the preferred embodiment for FISH the assay is performed on glass slides designed to hold and separate several samples.
Another subject of the present invention is a hybridisation method comprising    (a) contacting at least one nucleic acid of any of the present invention or a combination of nucleic acids of the present invention with a biological sample,    (b) hybridising the nucleic acid or the combination of nucleic acid of (a) with the sample under conditions where the stem of the nucleic is open, e.g. hybridising with a buffer which is essentially free of Mg2+, and    (c) inducing conditions which allow for stem formation in those nucleic acid molecules of (a) not forming a hybrid with the sample, e.g. washing with a Magnesium containing buffer, for instance at pH>8 or/and at room temperature.
The sample may be any sample of biological origin, such as a clinical or food sample, suspected of comprising a nucleic acid to be detected by the beacon. The sample may be a sample comprising microorganisms, such as bacteria, yeasts and molds, in particular Gram positive or/and Gram negative bacteria.
Also employed in the hybridisation method of the present invention can be a kit or chip as described herein.
“Essentially free of Mg2+” refers to a Mg2+ concentration of less than 1 mM, preferably less than 0.1 mM, more preferably less than 0.05 mM, most preferably less than 0.01 mM.
The buffer in step (c) may contain about 1 to about 20 mM Mg2+, more particular about 5 to about 15 mM Mg2+, even more particular about 8 to about 12 mM Mg2+, most particular about 10 mM Mg2+.
Any suitable hybridisation protocol comprising application of an essentially Mg2+ free solution and a Mg2+ containing solution as indicated above may be applied. For instance, the following protocol may be used: Aliquots of clinical samples are applied to defined fields on the slides. Preferably a defined quantity of 10 μl is applied and dried.    1. The samples are the heat fixed to the slides.    2. Gram positive organisms are subjected to a Lysozyme/Lysostaphin digestion following well published specifications. In a preferred embodiment the digestion is run for 7 minutes at 46° C. in a humidified chamber.    3. Pores are then formed for instance by immersing the slide 100% methanol or ethanol for several minutes. In a preferred embodiment the methanol or ethanol is ice cold and the immersion time is 7 minutes for Gram negative organisms and 3 minutes for Gram positive organisms.    4. The slide is then dried on a slide warmer, for instance at 55° C.    5. The beacons are dissolved in a hybridisation buffer (which may be essentially free of Mg2+) and then applied to each field of the slide while on the slide warmer.    6. The slide is placed in a hybridisation chamber, humidified with hybridisation buffer. In a preferred embodiment the slide is covered with a hydrophobic cover slip and placed on a covered slide warmer at 46° C. for 12 minutes.    7. The slide is then washed with a Magnesium containing buffer, for instance at pH>8 or/and at room temperature. The buffer main contain about 1 to about 20 mM Mg2+, more particular about 5 to about 15 mM Mg2+, even more particular about 8 to about 12 mM Mg2+, most particular 10 mM Mg2+    8. The slide is then dried and may be mounted with mounting fluid and can be read under an epifluorescence microscope at a total magnification of for instance 400×, 600×, or 1000×.
Should other vessels be used for the hybridisation, the detection may be via flow-cytometry or automated fluorescence reader well known in the art.
Yet another embodiment of the present invention relates to Chip applications of the beacons of the present invention. For Chip applications the beacons need to be covalently attached to a carrier surface. To facilitate this, the 3′-terminal base of the designed beacons may be either biotinylated or linked via a hetero-bifunctional reagent to an enzyme using methods well known in the art of protein and nucleic acid chemistry. Biotinylated beacons may then be added to Streptavidin coated chips as can be obtained freely from commercial sources (19). In this application the respective biotinylated hairpin loops can be attached to plurality of distinct fields of one chip, for instance at least 10, at least 50, at least 100, at least 200, or at least 500 fields, or at the maximum 500, at the maximum 400 or at the maximum 300 fields. Total RNA can be extracted from samples using commercially available kits (20) and can be applied to the chip under hybridising conditions. After hybridisation the chip can be briefly washed with a magnesium containing buffer, for instance at pH>8. Fluorescence on a field marks the presence of specific target sequence, for instance a specific RNA indicating the presence of a respective organism in the sample.
In order to open hybridisation assays to large scale routine applications it is necessary to analyse a plurality of samples sequentially on one reusable chip. The design of the chip must allow large scale production, efficient quality control and long shelf live.
In order to meet these specifications, in another embodiment of the present invention, a beacon of the present invention is covalently attached to an enzyme exerting a signal by catalysing a specific reaction. In particular, the enzyme may exert an electrochemical signal. Suitable enzymes comprise, but are not limited to tyrosinase, peroxidase, sulfite oxidase, alkaline phosphatase, glucose oxydase, guanine oxidase. In a preferred embodiment the enzyme is recombinantly derived from a genomic sequence of a thermo- or hyperthermophylic organism to render it stable under hybridisation conditions and elevated temperatures (21). The enzyme may be attached to the beacon at one end of the beacon molecule. At the other end of the molecule, an inhibitor may be attached which is capable of inhibiting the enzyme activity. When no cognate sequence to said hairpin loops is present the inhibitor inhibits the enzyme and no signal is generated. In the presence of a cognate sequence the loop will remain unfolded with the inhibitor well removed from the enzyme and the enzyme will produce an electrochemical signal which can be detected by devices well described in the art. A linker may be employed for the attachment of the enzyme or/and the inhibitor, in particular for the attachment of the inhibitor.
In a further preferred embodiment glucose oxidase is attached to one end of the said hairpin loops and a glucose oxidase inhibitor, such as an adenine nucleotide or adenine nucleotide analogue is attached to the other end of the hairpin loop. Adenine nucleotides are known inhibitors of glucose oxidase (22, 23). A linker may be employed for the attachment of the glucose oxidase or/and the glucose oxidase inhibitor, in particular for the attachment of the glucose oxidase inhibitor. When no cognate sequence to said hairpin loops is present the inhibitor, in particular the adenine nucleotide inhibits the enzyme and no signal is generated. In the presence of a cognate sequence the loop will remain unfolded with the inhibitor well removed from the enzyme and the enzyme will produce an electrochemical signal which can be detected by devices well described in the art.
To perform such an assay a large plurality of sequences with identical characteristics (Table I) have been developed, which may be applied to defined positions on the detecting device (chip) respectively. Total RNA is extracted from a sample utilising extraction procedure and kits readily available on the market (20) and placed on the chip under hybridisation conditions. After the hybridisation the chip is washed with substrate buffer at 46° C. and the signal is read. At the end of the cycle all hybridised RNA is washed off with hybridisation buffer at elevated temperature. Preferably the wash temperature is chosen 10° C. above the respective Tm. In a preferred embodiment the chip is washed at 60° C. with hybridisation buffer. The temperature may then dropped to 46° C. to equilibrate for the next analytical cycle.